SMC Quarterly News
www.standmgt.org
Stand Management Cooperative
School of Forest Resources, University of Washington
1st Quarter 2012
From the Director
I am happy to take over for Dave Briggs as the Director of SMC
with the start of the New Year. I think I speak for everyone in
thanking Dave for his years of service to SMC. I was fortunate to
meet with some of you at Dave’s retirement party and I think we
were all moved by the outpouring of affection from all who have
worked with Dave. I have just begun my work in understanding the
finances and getting to know the needs of SMC’s members. My plan
is to meet with some of you in the first weeks of February, prior to
our strategic planning meeting the morning of February 15th. I hope
to meet with most of you within the next couple of months. I
appreciate Mike Mosman’s offer to host the meeting at the Port
Blakely Tumwater office. I will circulate an agenda in the coming
week.
Greg Ettl
SMC Director
In this issue we highlight Paul Footen’s results from a long-term
study of the carryover effect of forest fertilization. The work
demonstrates an important justification for maintaining a long-term
plot network, as without the repeated measurements we would be
unaware of the potential for fertilization to effect a second rotation.
The carryover effects on growth are not as large as from initial
fertilization but 2nd rotation fertilized stands were about 1 year
ahead in growth a decade after establishment. The information
should be useful to managers in determining the potential of fertilization to contribute to future return on investment. You will also
find descriptions of ongoing research by Kim Littke, Kevin Ceder, Nai
Saetern, Austin Himes and Jed Bryce.
inside:
From the Director
1
Thanks from Dave
1
ORGANON Update
2
Student Reports
2
Feature Article
6
Abstracts and Pubs
13
SMC Spring Meeting
Date Changed to
April 19, 2012, 8:30-4:00
Gifford Pinchot National
Forest Headquarters,
Vancouver, WA.
GP Headquarters, Vancouver, WA
D. Briggs Thanks You
I am now in my third week of retirement but busy as ever; just a different agenda. I
want to thank all of you for your advice and support over my 15 years as
SMC Director. A special thanks to the staff, project leaders, and those
who served as chair of the Policy Committee for all of their help. It was a
journey that I thoroughly enjoyed. I always approach every day with two
objectives; to have fun and make a difference, and during my term, the
SMC achieved many accomplishments and we generally had fun on the
way to get them done.
I also want to thank you for the Retirement Party on December 2. My
wife, Anne, son Jeremy, and I were overwhelmed by such a large turnout.
Dave and Anne Briggs
Establishment of the David Briggs Scholarship in Forest Management is
Dave’s retirement party
an incredible honor and I am especially pleased because it will make a
difference for students in the future. Again, I thank all of you so much. I will continue to be active in my interest area as I wrap up a number of things that are on
the back burner and hope to see you in the future.
Revision to ORGANON 9.1 and Using the
ORGANON DLLs in R
Dave Hann has modified the ORGANON (http://www.cof.orst.edu/cof/fr/
research/organon/) volume and taper equations to incorporate a revised
equation for predicting stump diameter of Douglas-fir. Therefore this change will
affect estimates of cubic foot and board foot volume. I have revised the ORGANON web site such that it now includes revised console version and revised
DLLs (both executables and source code).
Dave also added to the ORGANON web site the procedures developed by Dr.
Peter Gould of the Pacific Northwest Research Station, Olympia, WA and Dr.
David Marshall of Weyerhaeuser Company for using the ORGANON DLLs in
the R statistical package.
Meeting Announcement
Coastal Silviculture Meeting
Vancouver Island univeristy, Nanaimo, on Feb 22nd. The topic is “Intensive Silviculture –
Miracle or Myth”. http://www.coastalsilviculturecommittee.com/Winter%202012/Home%20page.htm
2
SMC Student Rersearch Updates
Type V Paired-Tree Project
Kim Littke and Rob Harrison
The Type V paired-tree project is starting the fifth year of measurements. Sixty-two
paired-tree installations have been measured for two-year growth and fertilizer
response. The last eleven installations will be measured after two years of growth in
the fall of 2012. Soil characteristics and nutrients have been sampled from all 72
Douglas-fir installations and 2 ponderosa pine installations. Soil moisture and temperature is being continuously sampled at all installations. Kim Littke is analyzing
data and writing her dissertation using the first sixty Douglas-fir installations to
understand the best predictors of soil water and nitrogen availability, foliar N
concentration and area, and Douglas-fir productivity. Also, mapped and measured
climate, water availability, nitrogen availability, foliage, and productivity variables are
being examined as predictors of current basal area, height, and volume growth and
fertilizer response.
Silvicultural Manipulation Consequences Report
SMC Type III Analysis
Kevin Ceder and Eric Turnblom
Work continues on the analysis of the Type III portion of the larger Silvicultural
Manipulation Consequences in Stand Management Cooperative sites TechTransfer project, a.k.a., the (SMC)2 report. The primary focus of these SMC Type
III installations is determining the true effects of planting density (100, 200, 300,
440, 680 or 1210 TPA) on growth and yield. Results presented at the 2011 spring
meeting showed significant effects of planting density in preliminary yield models.
While the design of the Type III installations was geared to test significant
differences of yields at ages between different planting densities, actual planting
densities ranged considerably around the target densities. This coupled with the
timing of installation establishment and measurement cycle length caused
different sets of site classes to be measured at each age. Accounting for these
factors in conventional, experimental design-type statistical tests requires a large
amount of work with some loss of information relative to using functional
models to determine significance of planting density. The resulting models can
be used to assess differences in yields given different planting densities and to
construct yield tables. In the coming quarter we expect to publish a report
summarizing our findings and presenting a set of yield models to predict per-acre
stand-level yields including basal area, quadratic mean diameter, top height, total
and merchantable cubic-foot volume, and board-foot volume. After the report
we will build the models into software, as a web-based calculator or stand-alone
application, allowing managers to easily use our models.
3
Silvicultural Manipulation Consequences Report
SMC Type I Analysis
Nai Saetern, Eric Turnblom, and Dave Briggs
Analysis is progressing on the SMC Type I portion of the larger Silvicultural Manipulation Consequences in Stand Management Cooperative sites Tech-Transfer
project, a.k.a., the (SMC)2 report. Regression models for the SMC Type I Stands
have been constructed to predict per-acre stand-level yields of diameter at breast
height (DBH), height, and board-foot volume. In these models DBH, height, and
board-foot volume were individually modeled as a function of age, initial trees per
acre, relative density, thinned or not thinned, site index, latitude, longitude, and
elevation. Mortality rates were also determined for stands that were thinned at
least once versus stands that did not receive any thinning.
The three models (DBH, height, and board-foot volume) share similar predictor
variables that significantly affect their response. To summarize broadly, the three
near-final models show that all predictors are significant, except for latitude in the
board-foot volume yield model. For board-foot volume, whether or not the stand
has been thinned seems to be the main determinant of yield. The final mortality
model shows that there is significantly lower % mortality in thinned stands compared to unthinned stands.
Currently, the models are being fine-tuned and summarized for a Master’s thesis.
Breast Height Branches as Indicators of Tree and Log Quality:
SMC Type III Installations
Jed Bryce and Eric Turnblom
As branches grow and develop they become knots in wood, which are the primary
structural features that diminish quality of lumber and remanufactured wood
products. Recent decades have seen a decline in the quality of lumber in part due
to lower density planting that promotes greater branching and crown development. Due to the lengthy time investment required to measure branches on the
entire stem, the SMC is studying the diameter of the largest limb at breast height
(DLLBH) as well as the number of branches in this whorl as an indication of
overall wood quality. This is a quick, nondestructive measurement that could
potentially be incorporated in a timber cruise. The current research focuses on
the Type III installations in which the primary areas of interest are the effects of six
different (100, 200, 300, 440, 680 and 1210 TPA) planting densities on growth and
yield. Research presented at the 2011 spring SMC meeting confirmed previous
studies on the significance of density on branch development. Current analysis
involves developing fixed- and mixed-effects models to predict the DLLBH, the
number of branches in this whorl and the age at which the branches die. The final
goal is to provide managers with web-based tools to forecast branch growth and
development accurately based on the above-mentioned planting densities.
4
The Fate and Uptake of Enhanced Efficiency Urea Fertilizers in Douglas-fir Plantations of Western Oregon and Washington
Austin Himes, Kim Littke, Rob Harrison, Betsy Vance, Brian Strahm,
Tom Fox and Jose Zerpa
Nitrogen fertilizer is commonly applied to intensively managed Douglas-fir stands as an
investment in increased yields. However, recent fluctuations in the price of urea have
demonstrated the possibility that Douglas-fir fertilization may not be cost effective in the
future. Furthermore, concerns about water pollution have brought fertilizer practices into
question. Increasing the efficiency of nitrogen delivery to target trees reduces leaching
losses that contaminate water and potentially increases productivity per unit of fertilizer
applied. Nitrogen uptake and use efficiency have been increased in agricultural systems
using “coated” or “enhanced efficiency” urea formulations. To explore the viability of some
enhanced efficiency urea formulas in Douglas-fir plantations we are applying 15N labeled
fertilizer and utilizing stable isotope techniques to track the uptake and fate of fertilizer
nitrogen. Four treatments of 15N labeled fertilizer are being applied at ten SMC sites in
western Washington and Oregon selected to represent the range of conditions in the
region. The four treatments are urea, urea coated with NBPT (Agrotainâ), polymer coated
urea (ESN by Agrium Advanced Technologies), and coated urea fertilizer (Arboriteâ CUF). By
tracking the 15N label in samples of different ecosystem components through one year, we
will determine the performance of these four treatments in terms of uptake efficiency,
system losses, and ecosystem partitioning. Labeled fertilizer was applied to five sites in the
spring of 2011, and will be applied to the last five sites in the spring of 2012. By the summer of 2012 analysis of system losses and uptake efficiency of the first five installations will
be complete.
Risk Assessment for Projected Nitrogen Lost From Different
Harvesting Intensities in Pacific Northwest Douglas-fir
Rob Harrison, Austin Himes, Eric Turnblom, Patrick Wauters, David
Briggs, W. Devine, I. Eastin, E. O’Neil, Paul Footen, Erica Knight and
Betsy Vance
The growth of seventy-two intensively managed, mid-rotation, Douglas-fir stands in western Oregon, Washington, and British Columbia will be projected to 50 years of age using
the SMC variation of the ORGANON growth and yield simulator. From the ORGANON
output, component biomass removal will be estimated for a standard bole-only harvest and
a more intense biomass removal treatment. Utilizing allometric equations for the region
and existing SMC data on total site nitrogen and foliar nitrogen of trees in the 72 stands,
we will estimate nitrogen removal under the two harvest intensities as a proportion of
total site nitrogen. Based on the proportion of total site nitrogen removed we can assess
the risk associated with different harvest intensities on the long-term nitrogen availability
of the region. This broad assessment should assist managers in future decisions about
biomass removal. Currently we are simulating the stands’ growth in ORGANON and
researching the best allometric equations to use for component biomass and nitrogen
partitioning of Douglas-fir for the Pacific Northwest. We hope to present a region wide
risk assessment by the summer of 2012.
5
Feature Article
The Carryover Effect: Assessments of the Continued Effects of
Nitrogen Fertilization of a Previous Stand on the Productivity of a
Subsequent Douglas-fir Stand in the Pacific Northwest
Paul W. Footen, Bob Gonyea, Bert Hasselberg, Gage Wagoner,
Brian Strahm, Royce Anderson, Natalie Schmidt, Dave Briggs,
Eric Turnblom and Rob Harrison
Introduction and Methods
Increases in nutrients and water holding capacity suggest fertilization treatments
could have even long-term effects on forest productivity, however few studies have
assessed the long-term residual effects of fertilization on the productivity of a
subsequent stand replanted on the same site (i.e. fertilization “carryover effects”).
We studied the carryover effect at five western Washington sites: Pack Forest,
Coyle, Hank’s Lake, Simpson Log Yard and Camp Grisdale. Each site included at
least one (untreated) control plot and at least one adjacent plot fertilized repeatedly with urea (totaling 896-1120 kg N ha-1) (Table 1). The productivity of the five
study sites were measured in 2000, 2006 and 2008 (Figure 1), and seedling or tree
heights and diameter at breast height (DBH) were recorded in the study plots after
each growing season in the fall from 1997-2008 (except in 2004 and 2007). Tree
biomass estimates were calculated using equations from Bruce and DeMars (1974)
and biomass ratio equations from Jenkins et al. (2003). Aboveground C pools of the
tree and understory vegetation cover was found by multiplying the dry weight
biomass by 0.512 (Birdsey, 1992).
Current-year foliar samples were collected from each plot at the end of the growing season in the fall. Total N concentration was determined by dry combustion
(Perkin-Elmer CHN Analyzer Model 2400, Norwalk, CT). Mean tree needle biomass
was estimated using Jenkins et al. (2003) biomass ratio equations. Nitrogen content
for each sample was determined by multiplying the average N concentration by
needle biomass.
Soil pits were dug at plot center leaving one undisturbed face about one meter
wide and to at least one meter in depth of mineral soil. On two of the plots a 50
cm depth could only be reached due to highly compacted glacial till. Because 50 cm
was the shallowest depth, only the 0-50 cm depth was considered in statistical
comparisons of all soil pits. Samples were then ground to <1 mm and total C and N
concentration was determined by dry combustion (Perkin-Elmer CHN Analyzer
Model 2400, Norwalk, CT). The equations used to calculate C and N content per
hectare were:
Mg C ha-1 = (mg C / g soil)(g soil / cm3 soil)(cm soil / 1)(Mg C / 109 mg C)(108 cm2 / ha)
kg N ha-1 = (mg N / g soil)(g soil / cm3 soil)(cm soil / 1)(kg N / 106 mg N)(108 cm2 / ha)
6
Forest floor samples were collected from the same locations as the soil pits
using a 0.25 m2 (0.5 x 0.5 m) sample frame. Dried material was ground to <1
mm and a subsample analyzed for nutrient content by dry combustion (PerkinElmer CHN Analyzer Model 2400, Norwalk, CT). Plot level C and N content was
calculated by multiplying concentration levels by biomass using the following
equation: [Nutrient concentration (mg/kg) * 1 g / 1000 mg * O horizon weight
(g) / 0.25 m2 * 10000 m2 / 1 ha * 1 kg / 1000 g]. Mean tree needle biomass was
estimated using Jenkins et al. (2003) biomass ratio equation. Nitrogen content for
each sample was determined by multiplying the average N concentration by
needle biomass.
Results and Discussion
Mean tree height on carryover plots was significantly (p < 0.1) greater than the
unfertilized controls from 2001 to 2006 (Figure 2). In 2008, the year of the most
recent study, the carryover plots and the controls were not significantly different
(p = 0.2), though tree growth was 14% greater (Figure 2). Mean DBH was
significantly greater on the carryover plots than on the unfertilized controls in
2005, 2006 and 2008 (Figure 3). In 2008 mean DBH was 22% greater (p = 0.07)
on the carryover plots (Figure 3). The overall differences in tree growth between
the carryover plots and the controls has generally increased over the time in this
study causing the carryover stands to be at least one full growing season ahead
of the unfertilized controls. When measured in 2008 the 9-11 year old Douglasfir trees on the carryover plots contained a mean biomass of 3960 kg ha-1, which
was significantly greater (p = 0.04) than the controls. Mean tree C pools for
Douglas-fir trees were 33% (500 kg C ha-1) greater on the carryover plots than
the controls (Table 3).
The increases in tree growth on the carryover sites is comparable to growth
responses found after the initial application of fertilizer said to only last 5-10
years. This strongly demonstrates the ability of a subsequent stand of trees to
benefit from an application of fertilizer to a previous stand applied over 17-25
years ago, challenging the conventional wisdom of past studies (Figure 1).
Both mean understory biomass and N content were significantly greater on the
carryover plots than on the unfertilized controls. Mean understory biomass was
86% (p = 0.05) greater on the carryover plots when compared to the controls
(Figure 6). Mean understory N content, when compared to controls was 111%
(p = 0.04) greater on the carryover plots.
The total soil nitrogen and carbon content of the unfertilized controls was
consistently greater than the carryover plots. Mean soil N content of 0-50 cm
depth (including O horizon) on the control plots was 10% greater, but not
significantly different (p = 0.33) than the carryover plots (Figure 4).
Belowground soil carbon content on the control plots was 18% higher (p =
0.01) than on the previously fertilized carryover plots (Figure 5). Nearly all forest
C and N pools were measured in this study. However, due to time and budget
constraints C and N pools for roots and N content of bolewood and tree
branches were not measured. Douglas-fir trees and understory vegetation
comprise the total aboveground C pool and were 80% (11.8 Mg ha-1) greater in
the carryover plots than the unfertilized controls (Table 3). Nitrogen content of
7
Douglas-fir foliage and understory vegetation were measured to determine the
total aboveground N pool, which was 104% (352.5 kg N ha-1) greater on the
carryover plots than the controls (Table 2).
When C and N pools of the forest floor and mineral soil pits (to 50 cm) were
added to the aboveground pools, total forest pools could be estimated. Total
forest C pools on the controls were 10% (14.7 Mg C ha-1) greater than the
carryover plots (Table 3). Total forest N pools were 6% (271 kg N ha-1) greater
on the carryover plots than on the controls. The fact that the carryover plots
had 271 kg N ha-1 more N than the controls strongly supports the carryover
effect hypothesis (Table 2).
The carryover effect is not well known and few studies like this one exist.
Therefore, more research of carryover sites is necessary to properly assess
these long-term effects of N fertilization on second and even third rotation
Douglas-fir. The need to understand how fertilization can continue to influence
above and belowground C and N pools of a subsequent stands is important. The
ability to understand this phenomenon will help land managers and policy
makers make better decisions about how to manage forestlands for long-term
productivity and increased C sequestration. It is our hope that this study will
also influence policy makers regarding the importance of belowground C pools
on managed forestlands, and that they will consider these pools when calculating
C budgets in the future.
References
Birdsey, R.A. 1992. Carbon storage accumulation in United States forest
ecosystem. General Technical Report WO-59. US Department of
Agriculture, Forest Service, Global Change Research, Northeastern Forest
Experiment Station, Radnor, PA.
Bruce, D., and D.J. DeMars. 1974. Volume equations for second growth Douglasfir. United States Department of Agriculture. Research Note. PNW-239.
Jenkins, J.C., D.C. Chojnochky, L.S. Heath and R.A. Birsey. 2003. National-scale
biomass estimators for United States tree species. Forest Science. 49:12-15.
8
Table 1: Carryover study site descriptions
Figure 1: Timeline of the history of the carryover study highlighting the year (or range of years) when
important events occurred such as: stand establishments, fertilization applications, harvest of stands, dates
studies were conducted.
9
Figure 2: Mean tree height (m) of Douglas-fir on five carryover stands from beginning of the study in 1997 to
2008. Tree heights on carryover plots were significantly greater (p < 0.1) than the controls from 2001 to
2006. In 2008 mean height on carryover plots was 14% (p = 0.2) greater than the controls. Note that the xaxis is offset slightly to allow differentiation of treatment mean data points.
Figure 3: Mean diameter (cm) at breast height (DBH) of Douglas-fir trees on five carryover stands from
beginning of the study in 1997 to 2008. DBH on carryover plots was significantly greater (p < 0.1) than
controls in 2005, 2006 and 2008. DBH was 22% (p = 0.07) greater than controls in 2008. Note that the xaxis is offset slightly to allow differentiation of treatment mean data points.
10
-1
Figure 4: Mean soil nitrogen (kg N ha ) for 0-50 cm of mineral soil including O horizon on five carryover
sites measured in 2008. Mean soil N content was 10% (p = 0.33) greater on control plots than on carryover
plots. However, these differences were not significant.
-1
Figure 5: Mean soil carbon (Mg C ha ) for 0-50 cm of mineral soil including O horizon on five carryover sites
measured in 2008. The control plots were significantly greater (p < 0.1) than the carryover plots.
11
-1
Table 2: Total above- and belowground nitrogen pools (kg N ha ) of five carryover study sites assessed in
2008.
-1
Table 3: Total above- and belowground carbon pools (Mg C ha ) of five carryover study sites assessed in
2008.
12
Abstracts and Publications
Sooyoung Kim;Thomas Hinckley; David Briggs. Classifying individual
tree genera using stepwise cluster analysis based on height and intensity metrics derived from airborne laser scanner data. Remote Sensing of Environment (December 2011), 115 (12), pg. 3329-3342
http://ejournals2.scholarsportal.info/details.xqy?uri=/00344257/
v115i0012/3329_citgusdfalsd.xml
Abstract:
This paper evaluates the ability of small footprint, multiple return and pulsed
airborne scanner data to classify tree genera hierarchically using stepwise
cluster analysis. Leaf-on and leaf-off airborne scanner datasets obtained in the
Washington Park Arboretum, Seattle, Washington, USA were used for tree
genera classification. Parameters derived from structure and intensity data from
the leaf-on and leaf-off laser scanning datasets were compared to ground truth
data. Relative height percentiles and simple crown shapes using the ratio of a
crown length to width were computed for the structure variables. Selected
structure variables from the leaf-on dataset had higher classification rate
(74.9%) than those from the leaf-off dataset (50.2%) for distinguishing
deciduous from coniferous genera using linear discriminant functions.
Unsupervised stepwise cluster analysis was conducted to find groupings of
similar genera at consecutive steps using k-medoid algorithm. The three
stepwise cluster analyses using different seasonal laser scanning datasets
resulted in different outcomes, which imply that genera might be grouped
differently depending on the timing of the data collection. When combining leafon and leaf-off LIDAR datasets, the cluster analysis could separate the
deciduous genera from evergreen coniferous genera and could make further
separations between evergreen coniferous genera. When using the leaf-on
LIDAR dataset only, the cluster analysis did not separate deciduous from
evergreen genera. The overall results indicate the importance of the timing of
laser scanner data acquisition for tree genera separation and suggest that the
potential of combining two LIDAR datasets for improved classification.
Kantavichai, Rapeepan 2011. Effects of Climate and Thinning on
Coastal Douglas-fir Annual Biomass Growth at Four Sites. PhD
Dissertation, School of Forest Resources, University of Washington, Seattle, Wa. 148pp. Contact Megan O’Shea moshea@uw.edu
if you would like a copy.
Abstract:
Issues related to global warming and human releases of CO2, primarily from
the use of fossil fuels, have simulated interest in using forest biomass and
forest products to remove store carbon through photosynthesis which
removes atmospheric CO2 and using forest biomass for energy as a fossil
13
Abstracts and Publications cont.
fuel substitute. Unfortunately, current methods for estimating forest
biomass and the amount of carbon and energy it contains are crude and
inaccurate and relatively little research has been done to understand how
silviculture and climate affect a tree’s annual biomass growth and its
distribution within the tree. This study uses a unique set of x-ray densitometer data obtained from five positions along the stems of Douglas-fir
trees from a thinning trial on four sites in western Oregon and Washington to examine how thinning and climate affect the biomass increment
and its distribution. The x-ray densitometer data, combined with local
climate data over the life of each stand permitted examination of how
variations in climate at each site and imposition of thinning between 1987
and 1991 and harvested in 2006 at age 33-50. Not surprisingly responses
differed greatly between the sites due to great geographic dispersion and
soil differences which affect water deficit conditions. However, it was
found that, compared to unthinned controls, thinning increased annual
biomass increment. Immediately following thinning, there was greater ring
mass and relatively more latewood produced toward the lower stem and
relatively less toward the upper stem. This is consistent with stresses due
to wind sway and heavier crowns after thinning and the reduction of
competition for water by the residual stand. Although ring mass gains due
to thinning were still evident 12 years, later response was lower and there
was a reversal of the mass distribution patterns along the stem. Ring mass
production and distribution were also affected by climate. Higher summer
temperatures worsened drought and water stress conditions and lowered
ring mass growth; likely a reduction in growth when dense latewood
.
would
typically be produced. This did not occur on one site seemed to
always be in drought-induced summer dormancy. It was also apparent
that the an early return of cold winter-like storms in the fall, the continuation of such storms into the spring, or winter flooding can reduce ring
mass growth and alter the
14